Confusing data with theory

Maybe because experiments can be so much work, molecular biologists are just happy to have the data:

Krakauer, et al. “The challenges and scope of theoretical biology”, Journal of Theoretical Biology Volume 276, Issue 1, 7 May 2011, Pages 269–276:

The current absence of a strong theoretical foundation in biology means that there is weak guidance regarding what quantities or variables need to be understood to best inform a general understanding (an explanatory basis) for biological features of interest. An unfortunate result of the absence of theory is that some researchers confuse just having data with ‘understanding’. For example there is a base for collecting and analyzing the most microscopic data: experimental procedures and measurements in a high-throughput transcriptomics study are built around the assumption that transcripts are the primary data to be explained, and in neuroscience, recording from numerous individual neurons. This bias reflects a rather naive belief that the most fundamental data provide a form of explanation for a system, as if enumerating the fundamental particles were equivalent to the standard model in physics.

And here is this kind of thinking in action:

Nurse and Hayles. The cell in an era of systems biology. Cell (2011) vol. 144 (6) pp. 850-854:

The availability of ensemble datasets also allows the systematic grouping of genes with related functions. For example, catalogs of genes that when deleted have a similar cellular phenotype will identify gene sets required for particular processes… In this way, the ‘‘toolkit’’ required for a specific cellular process can be assembled. Another grouping approach is to construct networks based on gene products that interact with each other. Such networks can be assembled using interaction trap methodologies (such as two-hybrid methodologies and immunoprecipitations) that assess whether molecules are in physical contact. Also important are catalytic interactions resulting in metabolic changes or chemical modifications, such as phosphorylation… These various methodologies allow networks to be built up that connect molecular components throughout the cell to generate an overall cellular interactome…The power of these networks is enhanced when they are combined with catalogs of genes involved in a particular cellular function because they lead to a better molecular understanding of the process of interest…For many researchers, the creation of interaction networks is a major goal of systems cell biology that is aimed at providing complete networks of different cellular processes.

I read stuff like that, and I’m ready to throw in the towel.

Unknown's avatar

Author: Mike White

Genomes, Books, and Science Fiction

Leave a comment